Mathematical Money | October 1, 2026
$MCD
McDonald's is 32% below its February high and made a new low every single session last week. Plenty of people are asking whether this is the moment to step in.
I don't own it, I've been watching it, and my answer is not yet. Here's the reasoning, fundamental and technical, including the part where my own framework got caught out.
What actually happened on the 23rd
This wasn't a market selloff dragging a good company down. McDonald's fell 4.8% on the day of its own investor day, which is a fairly unusual way to lose money.
Management laid out a strategy called NEXT — roughly $8.5 billion of franchisee support through 2036, going into restaurant remodels, technology and operations, with targets for share gains in chicken and beverages and an operating margin in the low-to-mid 50s by 2030. There's nothing obviously wrong with any of that as a plan.
Two things landed badly. The CFO put the payback period on the spend at five to six years. And the company guided US comparable sales to be slightly negative in Q3.
So the message investors heard was: we're going to spend heavily for a decade, you'll see the return somewhere around 2032, and in the meantime the business is still shrinking at home. Six brokers cut their targets within days — JPMorgan to $260 from $280, TD Cowen to $270 from $282, Gordon Haskett to $285 from $315, BMO to $310 from $335, Evercore to $300 from $320, Citi to $305 from $310.
The fundamental problem isn't the plan
It's what sits underneath the comps, and this is the number I'd want anyone looking at this to understand.
Last quarter global same-store sales were up 1.3% and US same-store sales up 0.8%. Those look like small positives. But traffic declined and the comp was carried by a higher average check. Then July turned negative outright, and now Q3 is guided slightly negative too.
A comp held up by price while customers walk out is a different animal from a comp driven by more visits. The first has a ceiling — you can only raise prices so far before you accelerate the thing you're trying to offset. The second compounds. McDonald's has been running the first kind for a while, which is precisely why it needs an $8.5 billion plan.
The things that are genuinely good remain good. Global unit growth of 4.3% is strong, the international business is holding, the dividend has grown for 49 straight years, and the balance sheet is not in question. This is not a distressed company and I want to be clear about that.
What the chart says, and it says it loudly
McDonald's closed Tuesday at $230.94. It sits below its 20-day average, below its 50-day, below its 100-day, and below its 200-day — and the 200-day is more than 20% above the current price. It's 32.3% off the February high of $341.06 and it is sitting at the very bottom of its range: the lowest close in 260 sessions.
Here's the detail I find most telling, though. Sixty-day annualised volatility is about 21%. That's low.
A stock making new lows on low volatility isn't panicking. There's been no capitulation, no washout, no violent flush that clears out the sellers. It's a steady, orderly de-rating — the market calmly repricing the business lower, day after day. I'd honestly rather see a 9% crash on huge volume, because that at least tells you someone has given up. This looks like people quietly leaving.
Falling knives are a cliché. Falling knives with no panic in them are the ones that keep falling.
The argument that actually decides it for me
Forget the chart for a second.
At $230.94 the dividend yields roughly 3.2%. That's a genuinely decent yield from a company that has raised its payout for 49 consecutive years, and in a different environment it would be enough on its own.
But the ten-year Treasury is paying 5.24%, and the thirty-year recently touched its highest since 2002.
So the proposition is: accept 3.2%, take full equity risk, and wait five to six years for a turnaround plan to pay off — when government paper hands you 5.24% with no execution risk, no traffic problem and no new US president to prove themselves.
That is the comparison I keep coming back to, and it's not close. The yield that made McDonald's a defensive hold for twenty years doesn't do that job when the risk-free rate is above it. The whole category of "safe dividend compounder" gets repriced in a world where cash pays more, and I don't think that process has finished.
Where my own framework got caught
Worth admitting this part. I had McDonald's in my library with a dated milestone path, and the first gate was the September 23rd investor day, flagged red, needing the stock to hold somewhere in the $260 to $290 range to keep the recovery case alive.
It closed at $238.32 that day and it's lower now. The gate failed on schedule, which is the system doing its job — but the write-up itself was done on the 12th at $253, and the stock has drifted more than 8% below that since. A thesis written three weeks ago at a price 10% higher isn't a thesis any more, it's a historical document.
That's a reminder to me more than to anyone else: check the date on the analysis before you lean on it.
What would change my mind
Not the price. Cheaper isn't a reason, and it's the reason most people give.
Traffic. Not comps — traffic specifically, because comps can be propped up by pricing for several quarters while the underlying customer count bleeds. If guest counts stabilise in the US, the entire bear case weakens regardless of what the headline comp prints, and I'd be interested quickly.
The next real checkpoint is Q3 earnings on 22 October, which gives us August and September comps plus the first read on new management's execution. That's three weeks away. Nothing about this situation requires a decision before then, and that's rather the point — there is no scarcity of opportunity here, only the feeling of one.
Two things I'd like other views on.
Does anyone else explicitly compare a dividend yield against the ten-year before buying an income name? It sounds obvious written down, but I think a lot of us are still mentally pricing defensives against a 2% risk-free rate that stopped existing a while ago.
And on the $8.5 billion — is a five-to-six year payback a sign of management thinking properly long-term, or a sign they know the near-term is unfixable? I've read it the second way. I'd be genuinely interested if someone reads it the first way, because that's the version where this is a bargain.
Stop guessing. Start calculating.
Live to fight another day. 🤙
Comments